RLHF Specialist – Remote

About the role

The RLHF Specialist will design and operate reinforcement‑learning‑from‑human‑feedback pipelines to improve large language models. Working remotely, you will collaborate with ML engineers to boost model safety, factual accuracy, and alignment with human values.

Key responsibilities

• Generate high‑quality preference data by ranking model responses on helpfulness, honesty, and harmlessness.
• Design multi‑turn prompts that stress‑test reasoning and safety, and write chain‑of‑thought rationales for reward‑model training.
• Collaborate with engineers to analyse failure modes, identify data gaps, and propose data‑driven interventions.
• Develop and iterate annotation strategies, ensuring consistency across a global team.
• Probe models for biases, hallucinations, and vulnerabilities, documenting findings for optimisation.
• Maintain a personal benchmark set, regularly re‑evaluating new model versions.

Required profile

• Minimum 2 years experience in data annotation, model evaluation, computational linguistics, or AI trust & safety.
• Strong understanding of reinforcement‑learning concepts such as PPO, trust‑region methods, and reward hacking.
• Hands‑on experience fine‑tuning open‑source LLMs (e.g., Llama 2/3, Mistral, Gemma) with LoRA/QLoRA.
• Familiarity with constitutional AI or self‑alignment techniques and contributions to open‑source alignment libraries.

Required skills

• Python
• PyTorch, JAX, TensorFlow
• LoRA, QLoRA
• LabelBox, Scale AI, Snorkel
• AWS SageMaker, GCP Vertex AI
• PPO, Trust‑Region, Reward‑Hacking concepts
• TRL, Transformer Reinforcement Learning, Axolotl libraries